Research On Interaction Control Method Of Lower Limb Rehabilitation Robot Based On Human Physiological Signals

In recent years,the trend of population aging has gradually increased,and the elderly are extremely vulnerable to diseases due to the decline in the function of their own immune system.Among them,stroke is an extremely common disease,the patients often have symptoms such as numbness,weakness of the lower limbs,and hemiplegia,which seriously endanger the health of the patients.Therefore,rehabilitation treatment for stroke patients is imminent.The traditional method of rehabilitation medical treatment is one-to-one treatment by rehabilitation physicians,but there are problems such as large personnel consumption,long rehabilitation period,and limited effects.Compared with traditional rehabilitation methods,the lower limb rehabilitation robot has more flexibility,higher rehabilitation training efficiency,which reduced the burden of rehabilitation physicians.However,in the process of rehabilitation training,patients have active motion intention,and there will be a mismatch between the robot and the affected limb movement,which is easy to cause human-robot confrontation.Moreover,the current research on human-machine interaction control strategies for the lower limb rehabilitation robot is slightly weak,and the complexity of actual working conditions is rarely considered,resulting in some security risks in the process of rehabilitation training.This article carries out in-depth research on the above issues,and the main research contents are as follows:(1)Aiming at the problem of man-machine coupling dynamic modeling of human lower limb,the anatomical structure of human lower limb is analyzed and simplified reasonably.Furthermore,in the process of rehabilitation training,the interaction force between the affected limb and the lower limb rehabilitation robot cannot be ignored.Considering the friction between the robot and the patient,a nonlinear,time-varying,strong coupling,flexible and safe lower limb dynamics model is constructed.The establishment of human-machine coupling dynamic model of lower limb lays the foundation for the design of human-machine interaction controller.(2)Aiming at the problem of human lower limb multi-joint motion intention recognition,a mapping model between multi-channel surface Electromyography(s EMG)signals and continuous motion of lower limb multi-joint is established via collecting surface s EMG signals of lower limb related muscles,so as to effectively identify the continuous motion intention of lower limb multi-joints and provide reasonable and safe rehabilitation training trajectory for patients.Furthermore,the neuro-fuzzy corrector was used to correct the influence of individual differences and posture changes on the model online,so as to provide a safe and comfortable rehabilitation training environment for patients.(3)To solve the square root problem of time-varying matrix online,from the perspective of control and referring to the idea of integral control,the noise-suppression zeroing neural network model and the generally noise-suppression zeroing neural network model are proposed.Through theoretical analysis,the stability,global convergence and exponential convergence of the proposed model are strictly proved.Then,the simulation experiment verifies that the proposed model has good robustness,which provides an algorithm framework for the research of human-machine interaction control method.(4)Aiming at the design of the human-computer interaction controller of the lower limb rehabilitation robot,a zeroing neural network controller is proposed combined with human active motion intention.Furthermore,considering the interference of noise,a noise suppression zeroing neural network controller is proposed to improve the robustness of the controller.Moreover,to avoid the secondary damage of patients in the process of rehabilitation training,a model predictive controller based on zeroing neural network is proposed,which effectively restricts the motion range and motion rate of the lower limb rehabilitation robot,and provides a security guarantee for the rehabilitation training of patients.